Non-visual mirror Brightpath Tutors · no customer data
Move the Target, Move the Team
How do alternative rules change cost, individual outcomes and the behaviour rewarded?
Narrative sequence
- Current rule — Every fictional transaction receives a payout under the current rule.
- Replay candidate — The same stable transactions replay under one candidate rule.
- Separate outcomes — Total cost, new-customer reward and individual change separate.
- Winners and losers — An affordable aggregate opens into winners and losers.
- Adversarial records — Gaming and boundary cases enter as explicit adversarial records.
- Sensitivity — Threshold and leave-one-person sensitivity challenge the ranking.
- Decision matrix — Candidate schemes finish in a trade-off matrix; behaviour remains UNKNOWN.
| Evidence | Value | Status / meaning |
|---|---|---|
| Current total | R186,400 | replay |
| Candidate total | R201,900 | replay |
| People higher | 9 | distribution |
| People lower | 5 | distribution |
| Median change | +4.2% | distribution |
| Behaviour response | UNKNOWN | not observed |
Trivial baseline
Current fictional commission rule replayed over fixed historical transactions.
Sensitivity
Move thresholds, add adversarial transactions and remove one person at a time; behavioural response remains outside arithmetic replay.
Strongest counter-case
Fixed-behaviour replay cannot predict how people respond when incentives change.
Verdict
SURVIVES: arithmetic cost and distribution. UNTESTABLE: behavioural uplift before rollout.
Synthetic example. Deterministic fictional records; not customer data or achieved customer results.